In the U.S., approximately 30,600 infants per year are born before 28 weeks of gestation (40 weeks is term). These infants, termed Extremely Low Gestational Age Neonates (ELGANs), experience high morbidity and mortality: 20% of ELGANs admitted to an NICU die before discharge, 20% of survivors have severe and 20% moderate neurodevelopmental impairment (NDI). Perinatal care costs for these infants exceed $18 billion every year and account for approximately half of total hospital charges for newborn care. New approaches are needed to improve these outcomes. Recombinant erythropoietin (Epo) is a promising novel neuroprotective agent. It is widely available, affordable, and has been used safely in neonates to stimulate erythropoiesis. There are extensive preclinical data to support its use as a neuroprotective intervention: Epo decreases acute brain injury following hypoxia ischemia by decreasing inflammation, oxidative and excitotoxic injury which results in decreased apoptosis; Epo also promotes normal brain maturation by increasing neurogenesis, angiogenesis, and by protecting oligodendrocytes. We hypothesize that neonatal Epo treatment of ELGANs will decrease the combined outcome of death or severe NDI from 40% to 30% (primary outcome), or moderate NDI from 60% to 40% (secondary outcome) measured at 24-26 months post menstrual age (PMA). Our specific aims are to compare 376 Epo-treated with 376 control infants to determine: 1) whether Epo decreases the combined outcome of death or NDI at 24-26 months PMA (NDI is defined as the presence of: CP or Bayley III Scales of Infant and Toddler Development cognitive or motor scale < 70); 2) the short-, intermediate- and long-term safety of neonatal high dose Epo administration to ELGANs; 3) whether neonatal Epo treatment decreases serial measures of circulating inflammatory mediators, and biomarkers of brain injury; 4) whether Epo treatment improves brain structure (volume of gray matter, white matter and cerebellum, brain gyrification, and tract-based spatial statistics) at 36 weeks PMA as measured by MRI. In an exploratory aim, we will determine which MRI quantitative measures best predict neurodevelopment at 24-26 months PMA. We anticipate that Epo treatment of ELGANs will confer improved neurodevelopmental outcome at 24-26 months PMA compared to placebo, and will provide a much-needed therapy for this group of vulnerable infants. Furthermore, we anticipate that Epo treatment will be safe, will decrease biomarkers of brain injury and inflammation, and will be associated with less preterm brain injury as determined by MRI at 36 weeks PMA. The CCC will work closely with the linked DCC to accomplish the proposed goals. The CCC will provide the clinical leadership and support for all sites, and the DCC will provide the systems and oversight for data collection, management, quality control, operational support and data analyses for the monitoring and final reporting of the study.